/
AI Flashcards
Save to my account
Sign up
AI Flashcards
Artificial Intelligence Essentials
Study
1
Question
What is the definition of Artificial Intelligence?
Page 1
Answer
- Artificial Intelligence (AI) is a branch of computer science that aims to create intelligent machines capable of behaving like humans, thinking like humans, and making decisions. - It enables machines to exhibit human-based skills such as learning, reasoning, and problem-solving without needing pre-programmed instructions for every task.
2
Question
How does AI differ from traditional programming?
Page 1
Answer
- Traditional programming requires explicit instructions for every possible scenario, while AI uses algorithms that allow machines to learn and adapt on their own. - AI enables machines to perform tasks with their own intelligence, making decisions based on data rather than fixed rules.
3
Question
What are the main reasons to learn Artificial Intelligence?
Page 2
Answer
- AI helps solve real-world problems accurately, such as health issues, marketing, and traffic management. - It enables creation of personal virtual assistants like Siri or Google Assistant. - AI builds robots for risky environments and opens paths to new technologies and opportunities.
4
Question
What are the primary goals of Artificial Intelligence?
Page 2
Answer
- Replicate human intelligence in machines. - Solve knowledge-intensive tasks efficiently. - Create an intelligent connection between perception and action. - Build systems that perform human-like tasks such as proving theorems, playing chess, planning surgeries, or driving cars. - Develop systems that exhibit intelligent behavior, learn independently, and provide advice to users.
5
Question
What disciplines contribute to Artificial Intelligence?
Page 2
Answer
- Mathematics for logical foundations and algorithms. - Biology for understanding natural intelligence and neural processes. - Psychology for insights into human cognition and behavior. - Sociology for social implications and human interaction. - Computer Science for implementation and programming. - Neurons Study for modeling brain functions. - Statistics for data analysis and learning patterns.
6
Question
What are the key advantages of Artificial Intelligence?
Page 3
Answer
- High accuracy with fewer errors due to data-driven decisions. - High speed in processing and decision-making, e.g., beating chess champions. - High reliability for repetitive tasks with consistent performance. - Useful in risky areas like bomb defusing or deep-sea exploration. - Provides digital assistants for personalized services in e-commerce. - Enhances public utilities like self-driving cars and facial recognition for security.
7
Question
What are the main disadvantages of Artificial Intelligence?
Page 3
Answer
- High cost for hardware, software, and maintenance. - Limited to programmed tasks; cannot think creatively outside its training. - Lacks emotions and feelings, potentially leading to impersonal or harmful interactions. - Increases human dependency on machines, reducing mental capabilities. - No original creativity; cannot match human imagination.
8
Question
What significant event marked the maturation of AI in 1943?
Page 4
Answer
- Warren McCulloch and Walter Pitts proposed the first model of artificial neurons, laying the foundation for neural networks in AI.
9
Question
What is the Turing Test and its purpose?
Page 4
Answer
- The Turing Test, proposed by Alan Turing in 1950, checks if a machine can exhibit intelligent behavior equivalent to a human by fooling an interrogator in a conversation. - It evaluates a machine's ability to mimic human responses without relying on speech conversion.
10
Question
What was the first AI program created and its achievement?
Page 5
Answer
- In 1955, Allen Newell and Herbert A. Simon created the 'Logic Theorist,' the first AI program. - It proved 38 out of 52 mathematical theorems and found more elegant proofs for some.
11
Question
What event coined the term 'Artificial Intelligence' in 1956?
Page 5
Answer
- John McCarthy adopted the term 'Artificial Intelligence' at the Dartmouth Conference, establishing AI as an academic field. - This period saw the invention of high-level languages like FORTRAN, LISP, and COBOL, boosting AI enthusiasm.
12
Question
What was ELIZA and its significance in AI history?
Page 5
Answer
- ELIZA, created by Joseph Weizenbaum in 1966, was the first chatbot. - It demonstrated early natural language processing by simulating conversation, influencing AI's focus on human-machine interaction.
13
Question
What characterizes the first AI winter (1974-1980)?
Page 5
Answer
- It was a period of reduced funding and interest in AI research due to unmet expectations and resource shortages. - Publicity and government support for AI significantly decreased during this time.
14
Question
What marked the boom of AI in the 1980s?
Page 5
Answer
- The rise of Expert Systems that emulated human decision-making. - The first national conference of the American Association of Artificial Intelligence was held at Stanford in 1980.
15
Question
What was IBM Deep Blue's achievement in 1997?
Page 6
Answer
- IBM's Deep Blue defeated world chess champion Garry Kasparov, becoming the first computer to beat a human chess champion. - This highlighted AI's prowess in strategic games.
16
Question
What is Narrow AI and provide examples?
Page 7
Answer
- Narrow AI, or Weak AI, performs dedicated tasks intelligently but cannot operate beyond its specific training. - Examples: Apple's Siri for voice commands, IBM's Watson for question-answering, chess-playing programs, e-commerce recommendations, self-driving cars, speech and image recognition.
17
Question
What defines General AI?
Page 7
Answer
- General AI aims to perform any intellectual task with human-like efficiency and adaptability. - It would think and learn like humans across diverse domains, but no such system exists yet and requires extensive research.
18
Question
What are the characteristics of Super AI?
Page 7
Answer
- Super AI surpasses human intelligence, performing all tasks better with cognitive abilities like thinking, reasoning, learning, and communication independently. - It is a hypothetical outcome of General AI, still far from realization.
19
Question
What are Reactive Machines in AI functionality?
Page 8
Answer
- Reactive Machines are basic AI types that respond only to current stimuli without storing past experiences. - Examples: IBM's Deep Blue for chess and Google's AlphaGo for Go, focusing on present scenarios for optimal actions.
20
Question
How do Limited Memory AI systems work?
Page 9
Answer
- Limited Memory AI stores and uses recent data or experiences for short periods to inform decisions. - Example: Self-driving cars that track nearby vehicle speeds, distances, and speed limits for navigation.
21
Question
What is the Theory of Mind AI?
Page 9
Answer
- Theory of Mind AI would understand human emotions, beliefs, and social interactions to communicate like humans. - It remains under development, with ongoing research to achieve human-like empathy and social awareness.
22
Question
Describe Self-Awareness AI.
Page 9
Answer
- Self-Awareness AI would possess consciousness, sentiments, and self-awareness, making it super intelligent beyond human levels. - It is a hypothetical future concept that does not yet exist.
23
Question
What are the four major ethical issues in AI for healthcare?
Page 10
Answer
- Informed consent for using patient data. - Safety and transparency in AI algorithms. - Algorithmic fairness to avoid biases. - Data privacy to protect sensitive information.
24
Question
What is an AI agent?
Page 10
Answer
- An AI agent is an autonomous program or system that perceives its environment via sensors, makes decisions, and acts through actuators to achieve goals. - It operates without direct human control and can be reactive, proactive, or collaborative in single or multi-agent systems.
25
Question
What are the components of an AI agent's structure?
Page 11
Answer
- Architecture: The physical machinery with sensors and actuators, like a robot or computer. - Agent Program: Implements the agent function, mapping percept sequences to actions. - Agent = Architecture + Agent Program.
26
Question
Provide examples of AI agents in everyday applications.
Page 11
Answer
- Intelligent personal assistants like Siri or Alexa for tasks such as scheduling. - Autonomous robots like Roomba for cleaning. - Gaming agents for chess or poker. - Fraud detection in banking. - Traffic management in smart cities.
27
Question
What is a Simple Reflex Agent?
Page 13
Answer
- Simple Reflex Agents decide based solely on current percepts using condition-action rules, ignoring history. - They succeed in fully observable environments but lack adaptability and intelligence for complex or changing scenarios. - Example: A room cleaner that activates only when dirt is detected.
28
Question
How does a Model-based Reflex Agent differ from a Simple Reflex Agent?
Page 14
Answer
- Model-based Reflex Agents maintain an internal state and world model to handle partially observable environments. - They update based on how the world evolves and how actions affect it, enabling better tracking of situations. - This allows actions in environments where full observation isn't possible.
29
Question
What makes Goal-based Agents proactive?
Page 14
Answer
- Goal-based Agents use goal information to evaluate actions and plan sequences to achieve desirable outcomes. - They consider long-term scenarios through searching and planning, expanding beyond model-based capabilities. - This proactivity helps in complex decision-making where current state alone isn't enough.
30
Question
How do Utility-based Agents improve on Goal-based Agents?
Page 15
Answer
- Utility-based Agents measure success with a utility function, choosing the best path among multiple goal-achieving options. - They evaluate states with real numbers for efficiency, ideal for scenarios with trade-offs or alternatives. - This adds optimization for the most beneficial outcomes.